Cost-efficient routing, modulation, wavelength and port assignment using reinforcement learning in optical transport networks

•This paper carries out a detailed literature review about this topic, including resource assignment and reinforcement learning (RL)-based algorithm in OTN.•We explore multi-modal information in OTN, and express OTN and services through image.•This paper proposes a cost-efficient routing, modulation, wavelength, and port assignment (CE-RMWPA) algorithm based on RL to optimize the cost caused by ports.•The performance of CE-RMWPA is evaluated by simulation under different-scales of service. Simulation results show that the CE-RMWPA can reduce the cost.